臺灣博碩士論文加值系統

紡織品的表面瑕疵檢測，是為了確保最終產品品質之最重要過程。而傳統上之人工驗布是浪費時間及勞力密集的。因此，目前有許多研究應用現代電腦視覺技術於『織物瑕疵自動品質檢測系統』來解決這些問題。但是，這些研究大都是檢測梭織物之胚布或素色不織布之表面瑕疵。而蕾絲織物或是提花、印花織品等含有複雜紋路及圖案之瑕疵檢測，卻很少有相關之研究與應用。基於此種原因，本論文將應用影像處理技術來辨識蕾絲織物之表面瑕疵，並結合類神經網路理論，克服動態驗布時，因蕾絲織物橫移或旋轉所造成之辨識困難，以提昇織物瑕疵之分類與辨識率。
在本研究中，是分別以理論與實際蕾絲織物之瑕疵辨識分類結果作比較。首先，採用Area-Scan camera擷取理論與實際蕾絲織物瑕疵之影像。其次，將模擬蕾絲織物瑕疵之理論影像，作其瑕疵影像之平移及旋轉角度之動作，以模擬蕾絲織物在實際動態擷取影像所產生之各種影像變異，再分別利用影像處理之二值化處理，使其蕾絲織物之影像與背景之影像作區別。而為了能夠分辨出蕾絲織物之瑕疵特徵，本研究所採的方法為『不變矩』與『總物點數』之影像處理分析技術。並將所得到之瑕疵影像特徵值，組成一特徵向量作為類神經網路之輸入向量，並進行瑕疵分類。最後，再應用擷取實際蕾絲織物之瑕疵影像，來驗證之。由實驗的結果得知，理論蕾絲織物影像之瑕疵分類之正確率為100%，而實際蕾絲織物影像之瑕疵分類之正確率為93.33%。由此可知，本研究所採用之擷取瑕疵影像特徵值之方法，可以適用於蕾絲織物之『織物瑕疵自動品質檢測系統』中，以提高產品品質、降低提花織布製程與驗布上之成本，不僅能提昇出口競爭力外，更可提供更客觀之判別標準。

The inspection of fabric defects is the most important process for ensuring the quality of final products. Traditionally, manual inspections are not only time wasted but also labour intensive. To solve these problems, lots of researches apply updated computer visual technologies to the “Automatic Inspection of Fabric Defects System”. However, most of those researches focus on the inspection of woven fabrics or nonwoven fabrics. The researches and applications for the inspections of lace fabrics, jacquard fabrics, print fabrics or others are rare. Therefore, this study concentrates on improving the classification and recognizing accuracy of lace fabric defects by employing an image processing technology with an artificial neural network theory to solve the defect-recognition difficulty of lace fabrics caused by a slightly shift or a rotation of fabrics during a dynamic inspection process.
The approach of this study is to compare the defect-recognition results between theoretical lace fabric images and real lace fabric images. Firstly, Area-Scan CCD camera is applied to acquire theoretical and real images of lace fabrics defects. Secondly, the theoretical image of lace fabric is then shifted and rotated by the computer to simulate the variances of lace fabric images generated by a dynamic acquisition process. Then, the binary technology is employed to distinguish the differences between the images of lace fabrics and background. In order to tell the feature values of lace fabrics defects, this study utilizes the image processing analysis technologies of “moment invariants” and “total black pixel”. All the feature values of defect images are consisted into feature vectors as the input vectors of a neural network to classify the defects. Finally, the real images of lace fabrics can be acquired to verify the system. From the results, it tells that the accuracy of theoretical defect images classification of lace fabrics is 100% but the real defect images classification of lace fabrics is 93.33%. Therefore, it can be suggested that this application of acquiring features values of defect images is suitable for the “Automatic Inspection of Lace Fabric Defects System” to improve product quality and reduce the cost of lace fabric processing and inspection.

摘要..........................................iii
Abstract...................................... iv
誌謝........................................... v
表目錄.........................................ix
圖目錄..........................................x
第一章 緒論....................................1
1.1 研究動機與目的..............................1
1.2 文獻回顧....................................1
1.3 研究步驟....................................5
1.4 論文架構....................................5
第二章 實驗設備................................7
2.1 硬體設備....................................7
2.2 程式軟體....................................8
2.3 程式流程....................................9
第三章 數位影像處理技術.......................11
3.1 數位影像...................................11
3.2 數位影像處理之基本步驟.....................11
3.3 擷取影像特徵之原理.........................12
3.3.1 二值化..................................12
3.3.2 不變矩..................................13
3.3.3 總物點數................................14
第四章 類神經網路理論.........................16
4.1 類神經網路簡介.............................16
4.1.1 生物神經元模型..........................16
4.1.2 人工神經元..............................17
4.1.3 常用的轉換函數..........................18
4.1.4 類神經網路架構..........................20
4.1.5 類神經網路之運作過程....................21
4.2 倒傳遞類神經網路簡介.......................22
4.2.1 倒傳遞類神經網路之演算法................22
第五章 實驗及結果.............................26
5.1 實驗步驟與過程.............................26
5.2 蕾絲織物影像之各種角度及位移之七個不變矩關係.............................................30
5.2.1 實驗環境與參數..........................30
5.2.2 取樣....................................30
5.2.3 擷取特徵................................34
5.2.4 結果與討論..............................38
5.3 蕾絲織物影像之總物點數關係.................39
5.3.1 實驗環境與參數..........................39
5.3.2 取樣....................................39
5.3.3 擷取特徵................................43
5.3.4 結果與討論..............................44
5.4 蕾絲織物瑕疵分類...........................45
5.4.1 理論之蕾絲織物瑕疵.......................45
5.4.1.1 實驗環境與參數........................45
5.4.1.2 取樣..................................45
5.4.1.3 擷取特徵..............................47
5.4.1.4 訓練及測試............................47
5.4.1.5 辨識結果..............................52
5.4.1.6 討論..................................55
5.4.2 真實之蕾絲織物瑕疵.......................56
5.4.2.1 實驗環境與參數........................56
5.4.2.2 取樣..................................56
5.4.2.3 擷取特徵..............................58
5.4.2.4 訓練及測試............................58
5.4.2.5 辨識結果..............................61
5.4.2.6 討論..................................63
第六章 結論...................................64
參考文獻.......................................65
附錄
A: 影像旋轉角度及相對應位移之不變矩特徵值......67
B: 影像旋轉角度及位移之總物點數特徵值..........89
C: 理論之蕾絲織物之各種影像特徵值..............92
D: 理論之蕾絲織物之各種影像特徵值之正規化......99
E: 真實之蕾絲織物之各種影像特徵值..............106
F: 真實之蕾絲織物之各種影像特徵值之正規化......109
作者簡介.......................................112

1.C. Iype and T. R. Wan, “Nep Structure Identification and Analysis Using Image Processing Methods”, Textile Research Journal, Vol.68, No.6, pp.889-899(1998)
2.Bugao Xu and Ling Yu, “Determining Fiber Orientation Distribution in Nonwovens with Hough Transform Techniques”, Textile Research Journal, Vol.67, No.8, pp.563-571(1997)
3.F. Thorr, J. Y. Drean, and D. Adolphe, “Image Analysis Tools to Study Nonwovens”, Textile Research Journal, Vol.69, No.3, pp.162-168(1999)
4.J. Errol, Wood, “Applying Foruier and Associated Transforms to Pattern Characterization in Textiles”, Textile Research Journal, April, pp.212-220(1990)
5.J. Errol, Wood, “Description and Measurement of Carpet Appearance”, Textile Research Journal, Vol.63, No.10, pp.580-594(1993)
6.Bugao Xu, “Assessing Carpet Appearance Retention by Image Analysis”, Textile Research Journal, Vol.64, No.12, pp.697-709(1994)
7.W. Van Steenlandt, D. Collet, S. Sette, P. Bernard, R. Loning, L. Tezer, K. H. Bohland, and H. J. Schulz, “Automatic Assessment of Carpet Wear Using Image Analysis and Neural Networks”, Textile Research Journal, Vol.66, No.9, pp.555-561(1996)
8.Kenichi Ohta, Katsuhiko Sakaue, and Hideyuki Tamura, “Pattern Recognition of Fabric Surfaces”, Journal of the Textile Machinery Society of Japan, Vol.32, No.1,pp.7-10(1986)
9.Yixiang Frank Zhang, and Randall R. Bressee, “Fabric Defect Detection and Classification Using Image Analysis”, Textile Research Journal, Vol.65, No.1, pp.1-9(1995)
10.I-Shou Tsai, Chung-Hua Lin, and Jeng-Jong Lin, “Applying an Artificial Neural Network to Pattern Recognition in Fabric Defect”, Textile Research Journal, Vol.65, No.3, pp.123-130(1995)
11.S. A. Hosseini Ravandi, and K. Toriumi, “Fourier Transform Analysis of Plain Weave Fabric Appearance”, Textile Research Journal, Vol.65, No.11, pp.676-683(1995)
12.Warren J. Jasper, and Harsh Potlapalli, “Image Analysis of Mispicks in Woven Fabric”, Textile Research Journal, Vol.65, No.11, pp.683-692(1995)
13.Jeng-Jong Lin, Chung-Hua, and I-Shou Tsai, “Applying Expert System and Fuzzy Logic to an Intelligent Diagnosis System for Fabric Inspect”, Textile Research Journal, Vol.65, No.12, pp.697-709(1995)
14.I-Shou Tsai, and Ming-Chuan Hu, “Auto Inspection of Fabric Defects Using an Artificial Neural Network Technique”, Textile Research Journal, Vol.66, No.7, pp.474-482(1996)
15.Bugao Xu, “Identifying Fabric Structures with Fast Fourier Transform Techniques”, Textile Research Journal, Vol.66, No.8, pp.496-506(1996)
16.Chang Kyu Park, and Tae Jin Kang, “Obiective Rating of Seam Pucker Using Neural Networks”, Textile Research Journal, Vol.67, No.7, pp.494-502(1997)
17.Pei-Wen Chen, Tsair-Chun Liang, Hon-Fai Yau, Wan-Li Sun, Nai-Chueh Wang, Horng Chyi Lin, and Rong Cherng Lien, “Classifying Textile Faults with a Back-Propagation Neural Network Using Power Spectra”, Textile Research Journal, Vol.68, No.2, pp.121-126(1998)
18.S. Rajasekaran, “Training-Free Counter Propagation Neural Network for Pattern Recognition of Fabric Defects”, Textile Research Journal, Vol.67, No.6, pp.401-405(1997)
19.Tae Jin Kang, Chang Hoon Kim, and Kyung Wha Oh, “Automatic Recognition of Fabric Weave Patterns by Digital Image Analysis”, Textile Research Journal, Vol.69, No.2, pp.77-83(1999)
20.H. R. Yazdi, and T. G. King, “Application of ‘Vision in the Loop’ for Inspection of Lace Fabric”, Real-Time Imaging, Vol.4, pp.317-332(1998)
21.L. Vangheluwe, S. Sette, and F. Pynckels, “Assessment of Set Mark by Means of Neural Nets”, Textile Research Journal, Vol.63, No.4, pp.244-246(1993)
22.G. R. Barrett, T. G. Clapp, and K. J. Titus, “An On-Line Fabric Classification Technique Using a Wavelet-Based Neural Network Approach”, Textile Research Journal, Vol.66, No.8, pp.521-528(1996)
23.Kuo-Chin Fan, Yuan-Kai Wang, Bih-Lan Chang, Tzu-Po Wang, Chi-Hsiung Jou, and I-Feng Kao, “Fabric Classification Based on Recognition Using a Neural Network and Dimensionality Reduction”, Textile Research Journal, Vol.68, No.3, pp.179-185(1998)
24.張智星，MATLAB程式設計與應用，清蔚科技股份有限公司(2000)
25.繆紹綱，數位影像處理活用-Matlab，全華科技圖書股份有限公司(1999)
26.蒙以正，MATLAB5專業設計技巧，碁峰資訊股份有限公司(2000)
27.吳成柯、戴善榮、程湘君、雲立實，數位影像處理，儒林圖書有限公司(1996)
28.Ming-Kuei Hu, “Visual Pattern Recognition by Moment Invariants”, IRE TRANSACTIONS ON INFORMATION THEORY, Vol.8, pp.179-187(1962)
29.葉怡成，類神經網路模式應用與實作，儒林圖書有限公司(2000)
30.Chin-Teng Lin, and C.S. Georce Lee, Neural Fuzzy System, Prentice Hall(1999)